DESCRIPTION (Applicant's abstract): This project will continue a 25 year surveillance effort assessing the levels and patterns of substance use among American Indian adolescents attending reservation schools. Each year of the five-year project a representative sample of 2000 Indian youth living on will be given a comprehensive drug use survey in their school classroom. In addition, the survey will contain questions regarding violence, victimization and delinquent behaviors. These data will be used as the beginning phase of a long-term surveillance of these behaviors. The project will also, for the first time, determine rates of drug use, violence and victimization among American Indian students in the historic Indian lands in Oklahoma by surveying a representative sample of 1500 students each year. The purposes of the surveillance work are to observe changes over time, to accurately describe these domains, to provide insight into the nature of drug use, violence and victimization and to inform the efforts of those designing and evaluating intervention efforts. Hierarchical Linear Modeling will be use to provide unbiased estimates of rates and standard errors and to compare reservation Indian youth and Oklahoma Indian youth with rural American youth. Additionally, the project contains extensive scales assessing levels and patterns of risk and protective factors for drug use, violence and victimization. A series of Structural Equation Models will be used to study the relationships between these outcome behaviors and the predictors of social (family sanctions, peer drug associations) and emotional distress (anxiety, depression, alienation, anger) domains, and to assess moderating effects on these models of age, gender, family support, school adjustment and cultural identification. A final goal of the project is to develop a series of recommendations, based on project findings, for the design of drug, alcohol and violence prevention programs that will be effective specifically for American Indian youth.